[Bug target/102652] Unnecessary zeroing out of local ARM NEON arrays

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102652

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED
   Target Milestone|---                         |14.0

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
This is fixed in GCC 14.1.0:
```
bug:
.LFB3916:
        .cfi_startproc
        ldr     q24, [x1]
        cmlt    v27.16b, v24.16b, #0
        mov     v25.16b, v27.16b
        mov     v26.16b, v27.16b
        st4     {v24.16b - v27.16b}, [x0], 64
        ldr     q28, [x1, 16]
        cmlt    v31.16b, v28.16b, #0
        mov     v29.16b, v31.16b
        mov     v30.16b, v31.16b
        st4     {v28.16b - v31.16b}, [x0]
        ret
```

Only 4 total mov which is needed to duplicate v31 into v29 and v30 (and v27
into v25/v26) for the st4 case.